Author: Robert Hyatt
Date: 07:02:40 09/21/99
Go up one level in this thread
On September 20, 1999 at 23:48:47, Howard Exner wrote: >On September 20, 1999 at 22:29:16, KarinsDad wrote: > >>On September 20, 1999 at 18:36:42, Howard Exner wrote: >> >>>On September 20, 1999 at 14:42:27, Robert Hyatt wrote: >>> >>>> >>>>Here are a couple of examples: >>>> >>>> +---+---+---+---+---+---+---+---+ >>>> 8 | | | | | | | *K| | >>>> +---+---+---+---+---+---+---+---+ >>>> 7 | *P| *P| | | | *P| *P| *P| >>>> +---+---+---+---+---+---+---+---+ >>>> 6 | | | | | | | | | >>>> +---+---+---+---+---+---+---+---+ >>>> 5 | | | | | | | | | >>>> +---+---+---+---+---+---+---+---+ >>>> 4 | | | | | | | | | >>>> +---+---+---+---+---+---+---+---+ >>>> 3 | | | | | | | P | | >>>> +---+---+---+---+---+---+---+---+ >>>> 2 | P | P | P | | | | P | | >>>> +---+---+---+---+---+---+---+---+ >>>> 1 | | | | | | | K | | >>>> +---+---+---+---+---+---+---+---+ >>>> a b c d e f g h >>>> >>>> >>>>(fen 6k1/pp3ppp/8/8/8/6P1/PPP3P1/6K1/ w) >>>> >>>>old (16.18 crafty) evaluates this as -36 (black is better) because of >>>>those two doubled kingside pawns. new crafty says +.36 because white has >>>>a potential outside passer on the c file, black can't force a passer without >>>>the king helping. Better... >>> >>>Black can force a passer here without the king's help. >>> >>>Try the below artificial (and illegal) example on crafty. It will test >>>whether black can force a passer without the king's help. It has that same pawn >>>configuration on the king's side. >>> >>>1kb5/1p1p1ppp/1P1P4/8/8/1p1p2P1/1P1P2P1/K1B5 w >> >>Yes, black can get a passer. But cannot white also get a passer if black tries? >>Hence, the kingside is effectively even (not taking into account how far >>advanced the pawns are). > >Yes, black could force a promotion and then white promotes one move later. >I was just puzzled by the statement that "black could not force a passer without >the king helping". It does not matter much though because I think >Bob is on the right track anyway as the two doubled pawns pose a tougher >resistence than the two connected ones. These endings are tricky still. > somehwere in this thread I hope I qualified that statement properly, adding "without sacrificing a pawn and giving black a passer also". And the cute thing here is that I do _not_ have special code to handle the two doubled pawns. My 'candidate passer' code simply figures out that there is no way to make a passer by pushing the pawns in a safe way... >It's great to see the programmers , you, Bob and others, wrestle with >tweaking the endgame play. Just like a sports professional trys to >improve the weakest part of his game so to goes the battle with computers >and their achilles heal, the endgame. Years back we all know how outright >foolish computer endgame play was. Today it is a much more refined >endeavor as many of the old "blind spots" are gone. >> >>KarinsDad :) I agree. I have watched Crafty win _many_ endings vs GM players on ICC that it couldn't have touched years ago (nor could Cray Blitz have done as well if you factor out search depth)...
This page took 0 seconds to execute
Last modified: Thu, 15 Apr 21 08:11:13 -0700
Current Computer Chess Club Forums at Talkchess. This site by Sean Mintz.